1. Field of the Invention
This invention relates to the cloning and expression of the cellular receptor molecules that are capable of binding the TGF-.beta. supergene family of proteins. The invention further relates to methods of production of the isolated receptor molecules and their uses.
2. Description of Related Art
Following the initial purification and characterization of transforming growth factor-beta (TGF-.beta.) as a homodimeric, 25-Kd peptide (Frolik et al., Proc. Natl. Acad. Sci. USA, 80: 3676-3680 [1983]; Assoian et al., J. Biol. Chem., 258: 71 55-7160 [1983); Roberts et al., Biochemistry, 22: 5692-5698 [1983]), there has been an exponential increase in knowledge relating to this molecule. The cloning of TGF-.beta.1 and the resultant elucidation of its precursor structure (Derynck et al., Nature, 316: 701-705 [1985]) have led to the identification of at least four other forms of TGF-.beta. and the definition of a larger gene family comprising many other structurally related, but functionally distinct, regulatory proteins.
There are now many polypeptides that belong to -the TGF-.beta. supergene family by virtue of amino acid homologies, particularly with respect to the conservation of seven of the nine cysteine residues of TGF-.beta. among all known family members. These include the mammalian inhibins (Mason et al., Nature, 318: 659-663 [1985]) and activins (Ling et al., Nature, 321: 779-782 [1986]), and Mullerian inhibitory substance (MIS; Cate et al., Cell, 45: 685-698 [1986]), as well as the predicted products of both a pattern gene in Drosophila (the decapentaplegic gene complex, DPP-C; Padgett et al., Nature, 325: 81-84 [1987)), and an amphibian gene expressed in frog oocytes (Vgl; Weeks and Melton, Cell, 51: 861-867 [1987]). Most recently, three new proteins, called bone morphogenetic proteins (BMPS), have been added to the family. One subset of these proteins, BMP-2A and 2B, shares about 75% sequence identity with DPP-C and may represent the mammalian equivalent of that protein. Wozney et al., Science, 242: 1528-1534 (1988); Wang et al., Proc. Natl. Acad. Sci. USA, 85: 9484-9488 (1988).
In every case where the information is available, all polypeptides belonging to this family are encoded as larger precursors. The family resemblance is limited to the C-terminus of the precursor corresponding to the processed mature TGF-.beta. (Padgett et al., supra). With the exception of MIS, the C-terminal region is cleaved from the precursor at a pair of arginine residues. Although the position of this cleavage site varies among the family members, the C-terminus of all of the peptides is in the identical position, ending in the sequence where a cysteine residue is linked to the N-terminus of X-Cys-X (X being any amino acid), but differing in every case from the TGF-.beta. consensus C-terminal sequence where Cys is connected to the N-terminus of Lys-Cys-Ser.
A unifying feature of the biology of these polypeptides is their ability to regulate developmental processes. MIS induces regression of the female rudiments of the developing male reproductive system. The inhibins and activins, as discussed further below, regulate reproductive functions and erythropoietic activity. The BMPs are thought to play a role in the formation of cartilage and bone in vivo. DPP-C directs dorsal-ventral patterning in the developing fly embryo. Vgl is postulated to be involved in the process of induction of mesoderm from ectoderm during gastrulation in the amphibian embryo. In amphibians, TGF-.beta. itself (Kimelman and Kirschner, Cell, 51: 869-877 [1987]) has been shown to augment the ability of fibroblast growth factor to induce nesoderm and plays a pivotal role in morphogenesis and organogenesis in mammalian embryos. In addition, like activin, TGF-.beta. is reported to possess follicle-stimulating-hormone (FSH)-releasing activity. Ying et al., Biochem. Bioshys. Res. Commun., 135: 950-956 (1986).
Inhibin is a glycoprotein produced by diverse tissues, including the gonads, pituitary, brain, bone marrow, placenta, and adrenal gland. It was initially identified by its ability to inhibit the secretion of FSH by the pituitary. De Jong and Sharpe, Nature, 263: 71-72 (1976) ; Schwartz and Channing, Proc. Natl. Acad. Sci. USA, 74: 5721-5724 (1977). Such preferential regulation of the gonadotropin secretion has generated a great deal of interest and prompted many laboratories in the past fifty years to attempt to isolate and characterize this substance from extracts of testis, spermatozoa, rete testis fluid, seminal plasma, and ovarian follicular fluid using various bioassays. Rivier et al., Biochem. Bioshys. Res. Commun., 133: 120 (1985); Ling et al., Proc. Natl. Acad. Sci. USA, 82: 7217 (1985); Fukuda et al., Mol. Cell Endocrinol., 44: 55 (1985). The structure of inhibin, characterized from several species, consists of two disulfidelinked subunits: an .alpha. and either a .alpha..sub.A or a .beta..sub.B chain, designated "inhibin A" and "inhibin B," respectively.
After the identification of inhibin, activin was shown to exist in follicular fluid as a naturally occurring substance. Activin was found to be capable of stimulating FSH release by rat anterior pituitary cells. Vale et al., Nature, 321: 776-779 (1986) ; Ling et al., Nature, (1986), supra. Activin consists of a homodimer or heterodimer of inhibin P subunits, which may be .beta..sub.A or .beta..sub.B subunits. Vale et al., Recent Prog. Horm. Res., 44: 1-34 (1988). There is 95-100% amino acid conservation of .beta. subunits among human, porcine, bovine, and rat activins. The .beta..sub.A and .beta..sub.B subunits within a given species are about 64-70% homologous. The activin .beta..sub.A and .beta..sub.B homodimers ("activin A" and "activin B," respectively) have been identified in follicular fluid, and both molecules have been cloned and their genes expressed. Mason et al., Biochem. Biophys, Res. Commun., 135: 957 (1986) ; U.S. Pat. No. 4,798,885 issued Jan. 17, 1989; Mason et al., Molecular Endocrinol., 3:1352-1358 (1989). The complete sequence of the .beta..sub.B subunit is published in Serono Symposium Publications, InhibinNon-Steroidal Regulation of Follicle Stimulating Hormone Secretion, eds. H. G. Burger et al., abstract by A. J. Mason et al., vol. 42, pp. 77-88 (Raven Press: New York, 1987), entitled "Human Inhibin and Activin: Structure and Recombinant Expression in Mammalian Cells."
Both activin A and activin AB (the .beta..sub.A .beta..sub.B heterodimer), but thus far not activin B, have been isolated from natural sources. Activin MRNA (.beta..sub.A and .beta..sub.B subunits), bioactivity, and immunoactivity have been reported to be produced by testicular Leydig cells from immature rat and pig. Lee et al., Science, 243: 396-398 (1989); Lee et al., in Serono Symposium Publications, The Molecular and Cellular Endocrinology of the Testis, eds. Cooke and Sharpe, vol. 50, pp. 21-27 (Raven Press: New York, 1988). Activin A has been found recently to have erythropoietic-stimulating activity as well as FSH-releasing activity. EP Publ. No. 210,461 published Feb. 4, 1987 (where the protein is called BUF-3); Eto et al., Biochem. Biophys. Res. Commun., 142,. 1095-1103 (1987) Murata et al., Proc. Natl. Acad. Sci. U.S.A., 85: 2434-2438 (1988) (where the activin is called EDF) ; Yu et al., Nature, 330: 765-767 (1987) (where the activin is called FRP). In these systems, inhibin antagonized the actions of activin.
Recently, the expression of inhibin subunits, each encoded by a separate gene, was demonstrated in several tissues in addition to ovary and testis. Inhibin .alpha., .beta..sub.A, and .beta..sub.B mRNAs were detected in placental, pituitary, adrenal, bone marrow, and brain tissues. Meunier et al., Proc. Natl. Acad. Sci. USA, 85: 247-251 (1988). The expression of the inhibin subunit mRNAs varied by several-fold in a tissue-specific manner, suggesting different functions for these proteins depending on their pattern of association and their site of production.
In the human, growing preovulatory follicles and corpus luteum secrete inhibin into the circulation in response to FSH stimulation. Lee and Gibson, Aust, J. Biol, Sci., 38: 115-120 (1985); McLachlan et al., Fertil. Steril., 48: 1001 (1987). Thus, inhibin-related peptides play important roles in the modulation of gonadal functions via a pituitary feedback loop. In rat primary cultures of testis cells and ovarian thecal-interstitial cells, inhibin has been reported to enhance androgen biosynthesis stimulated by leutinizing hormone (LH), whereas activin suppresses androgen production. Hsueh et al., Prog. Natl. Acad. Sci, USA, 84: 5082-5086 (1987). Other workers have been unable to repeat these observations. dekretser and Robertson, Biology of Reproduction, 40: 33-47 (1989). Human ovarian theca were also shown to have a decrease in androgen production. It has now been shown that inhibin increases female fertility and activin decreases follicular size when they are administered locally to the ovaries. Woodruff et al., Endocrinol., 127: 3196-3205 (1990). On the other hand, activin was found to stimulate spermatogonial proliferation in germ-sertoli cell co-cultures from immature rat testis. Mather et al., Endocrinology, 127: 3206-3214 (1990).
It appears that each of the polypeptides in the TGF-.beta. supergene family has at least one unique receptor. Thus, although TGF-.beta.1 and TGF-.beta.2 compete for receptor binding (Cheifetz et al., Cell, 48: 409-415 [1987]; Segarini et al., J. Biol. Chem., 262: 14655-14662 [19871]), neither inhibin nor activin can compete for binding of TGF-.beta.1 to a variety of cell types, including pituitary cells (Cheifetz et al., J. Biol. Chem., 263: 16984-16991 [1988]), in which both TGF-.beta. and the activins elicit secretion of FSH, while inhibin antagonizes that activity. Ying et al., supra; Ying et al., Biochim Biophys. Res. Commun., 136: 969-975 [1986].
TGF-.beta. has been found to bind to nearly 150 different cell types and cell lines, and with only a few exceptions the cells bind TGF-.beta. with affinities in the picomolar range. Frolik et al., J. Biol. Chem., 259: 10995-11000 (1984); Tucker et al., Proc. Natl. Acad, Sci, USA, 81: 6757-6761 (1984); Massague and Like, J, Biol. Chem., 260: 2636-2645 (1985) ; Wakefield et al., J. Cell, Biol., 105: 965-975 (1987).
Crosslinking of labeled TGF-P to membrane receptors with disuccinimidyl suberate has revealed three distinct classes of integral cell membrane components that bind TGF-.beta. specifically and with high affinity. Massague, J. Biol. Chem., 20: 7059-7066 [1985]; Massague and Like, J. Biol. Chem., 260: 2636-2645 (1985); Cheifetz et al., J. Biol. Chem., 261: 9972-9978 (1986); Segarini and Seyedin, J. Biol. Chem., 263: 8366-8370 (1988); Cheifetz et al., J. Biol. Chem., supra; Cheifetz et al., J. Biol. Chem., 263: 10783-10789 (1988). Class I components are 65 Kd in all species, whereas class II components range from 85 Kd in rodent cells to 95 Kd in monkey and human cells to 110 Kd in chicken cells. The binding of the various forms of TGF-.beta. to both class I and II receptors is in the order TGF-.beta.1&gt;TGF-.beta.1.2&gt;TGF-.beta.2. Cheifetz et al., J. Biol. Chem., supra. In contrast, all three forms of TGF-.beta. bind equivalently to class III receptors, which represent the most abundant cross-linked species. This form is generally considered to be dimeric (Massague, J. Biol. Chem., 260: 7059-7066, supra) and to be composed of proteoglycan subunits of 250-350 Kd (Segarini and Seyedin, supra; Cheifetz et al., J. Biol. Chem., 16984-16991, supra). Another subset of this high-molecular-weight component has also been described that preferentially binds TGF-.beta.2. Segarini et al., J. Biol. Chem., 262: 14655-14662 (1987). Most frequently, all three classes of these binding proteins coexist on cells.
Class I and II TGF-.beta. receptor components, like most growth factor receptors, are glycoproteins. Most of the carbohydrate is N-linked and contributes approximately 5 Kd and 15-20 Kd to the mass of components I and II, respectively. Cheifetz et al., J. Biol. Chem., 16984-16991, supra. In contrast to all other known polypeptide receptors, the class III protein is a proteoglycan consisting predominantly of heparin sulfate glycosaminoglycan chains with a smaller amount of chondroitin or dermatan sulfate attached to a core protein of about 100-140 Kd. The binding site for TGF-.beta. resides in this core protein. Segarini and Seyedin, supra; Cheifetz et al., J. Biol. Chem., 16984-16991, supra.
There is considerable controversy concerning the roles of the various classes of proteins that cross-link to TGF-.beta.. Cheifetz and coworkers have proposed that class III receptors mediate al.1 functions of TGF-.beta. in which TGF-.beta.1 and TGF-.beta.2 have been shown to be equipotent; this includes regulation of extracellular matrix as well as most effects on growth and differentiation. Cheifetz et al., 1987 and 1988, supra. Moreover, they suggest that biological activities specific to TGF-.beta.1 are mediated through the class I and II receptors; these would include the reported selective inhibitory activity of TGF-.beta.1 on growth of either B6SUt-A multipotential hematopoietic progenitor cells (Cheifetz et al., J. Biol. Cbem., 10783-10789, supra) or endothelial cells. Jennings et al., J. Cell Physiol., 137: 167-172 (1988). In contrast, it has been shown that class III binding is not exhibited by primary epithelial, endothelial, and lymphoid cells and may not be necessary for many biological activities of TGF-.beta.. Segarini et al., Mol. Endocrin., 3: 261-272 (1989). Thus, cells such as L-6 myoblasts (Segarini et al., 1987, supra) and primary lymphocytes (Kehrl et al., J. Immunol., 143: 1868-1874 [1989)), which respond equally well to TGF-.beta.1 and TGF-.beta.2, have only class I and II receptors. In addition, preliminary data demonstrate that selection of cell mutants resistant to the action of TGF-.beta. results in the isolation of cell lines that have selectively lost the type I TGF-.beta. binding.
Although their occurrence is rather rare, several neoplastic cells appear to lack these putative TGF-.beta. receptors. These include the PC12 rat pheochromocytoma, human retinoblastoma cells, and several leukemic cell lines. Kimchi et al., Science, 240: 196-198 (1988); Keller et al., J. Cell. Biochem,, 39: 175-184 (1989). In each of these cases, the lack of receptor proteins correlates with resistance of the cells to the inhibitory effects of TGF-.beta., and it has been proposed that loss of TGF-.beta. receptors might be a mechanism whereby pre-neoplastic cells could progress to tumor cells by escaping from negative growth. Sporn and Roberts, Nature, 313: 745-747 (1985).
Yet another anomaly of TGF-.beta. binding in tumor cells has been reported. Although the binding of TGF-.beta. to cells has been shown to be specific, a novel 70-74 Kd complex has been reported on GH.sub.3 rat pituitary tumor cells that binds not only TGF-.beta.1, but also TGF-.beta.2, activin AB, and inhibin with lower affinity. Cheifetz et al., J. Biol. Chem., 263: 17225-17228 (1988). The biological function of this complex is not known.
Although it has long been known or suspected that the biological effects of inhibin and activin are mediated via interaction with a cellular receptor molecule present on the surface of target cells, to date such receptor(s) have never been isolated or identified. While a protein designated as an "activin receptor" has been expression cloned (Mathews and Vale, Cell, 65: 1-20 [1991]), it does not appear to be a mammalian receptor in that it is a serine kinase (not found in mammalian cells), and it is most closely related to the C. elegans daf-1 gene product (found in nematodes).
It is an object of the present invention to isolate and specifically identify and purify the cellular receptors to which the TGF-.beta. supergene family is capable of binding in vivo.
A more specific object of the present invention is to identify and isolate the cellular receptors to which inhibin and/or activin binds in vivo.
It is another object to provide nucleic acid molecules encoding such receptors.
It is yet another object to provide derivatives and modified forms of the TGF-.beta. supergene family of receptors, including amino acid sequence variants and covalent derivatives thereof.
These and other objects of the invention will be apparent to the ordinary artisan upon consideration of the specification as a whole.